Anti-ship missile formation recognition target selection system based on Hough transformation and optimized K-means clustering

Abstract

The invention discloses an anti-ship missile formation recognition target selection system based on Hough transformation and optimized K-means clustering. The anti-ship missile formation recognition target selection system comprises a motion situation monitoring module, a formation contour recognition module, a target finger situation establishment module, a striking target acquisition module and a target striking module. Based on the Hough transformation and the optimized K-means clustering algorithm, the anti-ship missile formation recognition target selection modeling method is provided, the efficiency is improved, the engineering operability is enhanced, and the method is of great significance to anti-ship combat simulation.

Description

technical field [0001] The invention relates to the field of formation recognition, in particular to an anti-ship missile formation recognition target selection system based on Hough transform and optimized K-means clustering. Background technique [0002] The target selection of anti-ship missiles usually has two types of technologies: feature recognition and formation recognition. Feature recognition relies on the single feature of the designated target that is different from other ships in the formation, or the comprehensive features weighted by multiple features. However, with the increase in the self-control time brought about by the increase in the range of anti-ship missiles, the changes in the situation of the terminal formation no longer meet the requirements determined before launch. characteristics, thereby exceeding the ability of anti-ship missile signature recognition. Formation recognition is to compare the formation situation captured by the missile terminal...

AI Technical Summary

Problems solved by technology

It is known that the requirements for formation sorting are too high, which reduces the applicability of the method, and the judgment degree of different formations is not high, and the alignment angle has too much influence on the judgment; 3) Establish a formation template with the number of queue lines and queue angles, The mathematical model of the reference target, other targets relative to the queue orientation and similarity measure is given to judge whether the target is in the same queue line, and the number of queue lines is obtained by clustering to identify the formation. The similarity measure requires that the azimuth angle deviation must be smaller than the detection error , the judgment of the target being in the same queue line is not rigorous enough

[0004] Another type of formation recognition is to convert the target from the measurement space to the parameter space parameters, combined with formation templates for matching judgment, such as establishing a feature model and feature template for the formation shape and the distribution of group members in the formation, based on the template sliding Matching and identifying formation types, this method can identify linear formations, but the clustering process is complicated and the real-time performance is weak; at the same time, there is a method of K-means clustering algorithm that optimizes the number of clusters with the D / L cost function, D / L The cost function has a decreasing characteristic, the clustering number optimization and the clustering iterative process are independent of each other, there are multiple local optimal values, the key clustering numbers are not clearly differentiated, and the optimization number for typical clustering problems is slightly higher than the theoretical value

The common problem is that the impact of the detection interval on the clustering effect is not considered

Images